Effect of Smoking on Treatment Efficacy and Toxicity in Patients with Cancer: A Systematic Review and Meta-Analysis

Abstract

Aim: The aim of the present systematic review and meta-analysis was to summarize the current evidence on the potential impact of smoking during cancer treatment on treatment efficacy and toxicity irrespective of cancer type.

Methods: A systematic literature search was performed using two electronic databases for potentially eligible studies. Only studies based on multivariable analysis for the association between smoking, compared to non-smokers (never or former), and treatment efficacy or toxicity were included. Pooled Hazard Ratios (HRs) or Odds Ratios (ORs) and corresponding 95% Confidence Intervals (CIs) were estimated through random-effects meta-analyses.

Results: In total, 97 eligible studies were identified, of which 79 were eligible for the pooled analyses. Smoking during radiation therapy, with or without chemotherapy, was associated with an increased risk of locoregional recurrence (pooled HR: 1.56; 95% CI: 1.28-1.91 for radiation therapy; pooled HR: 4.28; 95% CI: 2.06-8.90 for chemoradiotherapy) and worse disease-free survival (pooled HR: 1.88; 95% CI: 1.21-2.90 for radiation therapy; pooled HR: 1.92; 95% CI: 1.41-2.62 for chemoradiotherapy) as well as a higher risk for radiation-induced toxicity (pooled OR: 1.84; 95% CI: 1.32-2.56 for radiation therapy; pooled OR: 2.43; 95% CI: 1.43-4.07 for chemoradiotherapy) with low-to-moderate certainty of evidence. Smoking during treatment with EGFR tyrosine kinase inhibitors (EGFR-TKIs) in patients with lung cancer was associated with worse progression-free survival compared to non-smokers (pooled HR: 1.43; 95% CI: 1.14-1.80; moderate certainty of evidence), whereas smoking was associated with improved progression-free survival in patients treated with checkpoint inhibitors (HR: 0.70; 95% CI: 0.58-0.84; moderate certainty of evidence). No statistically significant associations were observed between smoking and treatment efficacy or toxicity to chemotherapy.

Conclusion: The present meta-analysis confirms earlier evidence of the negative impact of smoking during radiation therapy, with or without chemotherapy, on treatment efficacy and radiation-induced toxicity as well as a negative impact of smoking on the efficacy of EGFR-TKIs and a positive impact on the efficacy of checkpoint inhibitors. The evidence is too weak to draw firm conclusions on the potential association between smoking and chemotherapy, whereas there is no evidence for pooled analyses regarding other types of systemic oncological therapy.

Keywords: cancer; efficacy; meta-analysis; smoking; toxicity.

Figure 1

Flowchart diagram of study selection process.

Figure 2

Forest plots on pooled Hazard Ratios (HRs) on smoking during radiotherapy and efficacy. (a) Smoking during radiotherapy and locoregional recurrence free survival; (b) smoking during radiotherapy and disease-free survival. HR > 1 indicates worse outcome for smokers during radiation treatment, whereas HR < 1 indicates better outcome. Comparison group is non-smoker (former or never). Abbreviations: NSCLC, non-small-cell lung cancer; H&N, head and neck cancer.

Figure 3

Forest plots on pooled Hazard Ratios (HRs) on smoking during chemoradiotherapy and efficacy. (a) Smoking during chemoradiotherapy and locoregional recurrence free survival; (b) smoking during chemoradiotherapy and disease-free survival. HR > 1 indicates worse outcome for smokers during radiation treatment, whereas HR < 1 indicates better outcome. Comparison group is non-smoker (former or never). Abbreviation: H&N, head and neck cancer.

Figure 4

Forest plots on pooled Odds Ratios (ORs) on smoking during treatment and toxicity. (a) Smoking during radiotherapy and toxicity; (b) smoking during chemotherapy and toxicity. OR > 1 indicates higher risk for toxicity during treatment, whereas HR < 1 indicates lower risk. Comparison group is non-smokers (former or never). Abbreviations: NSCLC, non-small-cell lung cancer; H&N, head and neck cancer.